English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Real-world robotics problems often occur in domains that differ significantlyfrom the robot's prior training environment. For many robotic control tasks,real world experience is expensive to obtain, but data is easy to collect ineither an instrumented environment or in simulation. We propose a novel domainadaptation approach for robot perception that adapts visual representationslearned on a large easy-to-obtain source dataset (e.g. synthetic images) to atarget real-world domain, without requiring expensive manual data annotation ofreal world data before policy search. Supervised domain adaptation methodsminimize cross-domain differences using pairs of aligned images that containthe same object or scene in both the source and target domains, thus learning adomain-invariant representation. However, they require manual alignment of suchimage pairs. Fully unsupervised adaptation methods rely on minimizing thediscrepancy between the feature distributions across domains. We propose anovel, more powerful combination of both distribution and pairwise imagealignment, and remove the requirement for expensive annotation by using weaklyaligned pairs of images in the source and target domains. Focusing on adaptingfrom simulation to real world data using a PR2 robot, we evaluate our approachon a manipulation task and show that by using weakly paired images, our methodcompensates for domain shift more effectively than previous techniques,enabling better robot performance in the real world.

Adapting Deep Visuomotor Representations with Weak Pairwise Constraints